From: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
To: Andy Lutomirski <luto@kernel.org>,
Linus Torvalds <torvalds@linux-foundation.org>
Cc: David Howells <dhowells@redhat.com>,
Rasmus Villemoes <linux@rasmusvillemoes.dk>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Peter Zijlstra <peterz@infradead.org>,
Nicolas Dichtel <nicolas.dichtel@6wind.com>,
raven@themaw.net, Christian Brauner <christian@brauner.io>,
keyrings@vger.kernel.org, USB list <linux-usb@vger.kernel.org>,
linux-block <linux-block@vger.kernel.org>,
LSM List <linux-security-module@vger.kernel.org>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Linux API <linux-api@vger.kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Miklos Szeredi <miklos@szeredi.hu>
Subject: Re: [RFC PATCH 11/10] pipe: Add fsync() support [ver #2]
Date: Sun, 3 Nov 2019 15:02:05 +0300 [thread overview]
Message-ID: <c8001e7c-8039-3efb-948b-482b88005660@yandex-team.ru> (raw)
In-Reply-To: <CALCETrWZjW88OY2mh7v8cUU_6XTSJTkQhAfNbSC17AdhEWwVAA@mail.gmail.com>
On 03/11/2019 02.14, Andy Lutomirski wrote:
> On Sat, Nov 2, 2019 at 4:10 PM Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
>>
>> On Sat, Nov 2, 2019 at 4:02 PM Linus Torvalds
>> <torvalds@linux-foundation.org> wrote:
>>>
>>> But I don't think anybody actually _did_ any of that. But that's
>>> basically the argument for the three splice operations:
>>> write/vmsplice/splice(). Which one you use depends on the lifetime and
>>> the source of your data. write() is obviously for the copy case (the
>>> source data might not be stable), while splice() is for the "data from
>>> another source", and vmsplace() is "data is from stable data in my
>>> vm".
>>
>> Btw, it's really worth noting that "splice()" and friends are from a
>> more happy-go-lucky time when we were experimenting with new
>> interfaces, and in a day and age when people thought that interfaces
>> like "sendpage()" and zero-copy and playing games with the VM was a
>> great thing to do.
>
> I suppose a nicer interface might be:
>
>
> madvise(buf, len, MADV_STABILIZE);
>
> (MADV_STABILIZE is an imaginary operation that write protects the
> memory a la fork() but without the copying part.)
>
> vmsplice_safer(fd, ...);
>
> Where vmsplice_safer() is like vmsplice, except that it only works on
> write-protected pages. If you vmsplice_safer() some memory and then
> write to the memory, the pipe keeps the old copy.
>
> But this can all be done with memfd and splice, too, I think.
Looks monstrous. This will kill all fun and profit. =)
I think vmsplice should at least deprecate and ignore SPLICE_F_GIFT.
It almost never works - if page still mapped then page_count in
generic_pipe_buf_steal() will be at least 2 (pte and pipe gup).
But if user munmap vma between splicing and consuming (and page not
stuck in lazy tlb and per-cpu vectors) then page from anon lru
could be spliced into file. Ouch.
And looks like fuse device still accepts SPLICE_F_MOVE.
>
>
>>
>> It turns out that VM games are almost always more expensive than just
>> copying the data in the first place, but hey, people didn't know that,
>> and zero-copy was seen a big deal.
>>
>> The reality is that almost nobody uses splice and vmsplice at all, and
>> they have been a much bigger headache than they are worth. If I could
>> go back in time and not do them, I would. But there have been a few
>> very special uses that seem to actually like the interfaces.
>>
>> But it's entirely possible that we should kill vmsplice() (likely by
>> just implementing the semantics as "write()") because it's not common
>> enough to have the complexity.
>
> I think this is the right choice.
>
> FWIW, the openssl vmsplice() call looks dubious, but I suspect it's
> okay because it's vmsplicing to a netlink socket, and the kernel code
> on the other end won't read the data after it returns a response.
>
> --Andy
>
prev parent reply other threads:[~2019-11-03 12:02 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-23 20:17 [RFC PATCH 00/10] pipe: Notification queue preparation [ver #2] David Howells
2019-10-23 20:17 ` [RFC PATCH 01/10] pipe: Reduce #inclusion of pipe_fs_i.h " David Howells
2019-10-23 20:17 ` [RFC PATCH 02/10] Remove the nr_exclusive argument from __wake_up_sync_key() " David Howells
2019-10-23 20:17 ` [RFC PATCH 03/10] Add wake_up_interruptible_sync_poll_locked() " David Howells
2019-10-23 20:17 ` [RFC PATCH 04/10] pipe: Use head and tail pointers for the ring, not cursor and length " David Howells
2019-10-27 14:03 ` Linus Torvalds
2019-10-30 16:19 ` Ilya Dryomov
2019-10-30 20:35 ` Rasmus Villemoes
2019-10-30 22:16 ` Ilya Dryomov
2019-10-30 22:38 ` Rasmus Villemoes
2019-10-31 15:11 ` David Howells
2019-10-31 15:57 ` Ilya Dryomov
2019-11-01 14:53 ` David Howells
2019-10-31 14:57 ` David Howells
2019-11-03 11:17 ` Matthew Wilcox
2019-12-06 21:47 ` Johannes Hirte
2019-12-06 22:14 ` Linus Torvalds
2019-12-07 0:00 ` Johannes Hirte
2019-12-07 1:03 ` Linus Torvalds
2019-12-08 17:56 ` Johannes Hirte
2019-12-08 18:10 ` Linus Torvalds
2019-12-07 6:47 ` Linus Torvalds
2019-12-06 22:15 ` David Howells
2019-10-23 20:17 ` [RFC PATCH 05/10] pipe: Allow pipes to have kernel-reserved slots " David Howells
2019-10-23 20:18 ` [RFC PATCH 06/10] pipe: Advance tail pointer inside of wait spinlock in pipe_read() " David Howells
2019-10-23 20:18 ` [RFC PATCH 07/10] pipe: Conditionalise wakeup " David Howells
2019-10-27 15:57 ` Konstantin Khlebnikov
2019-10-31 15:21 ` David Howells
2019-10-31 16:38 ` David Howells
2019-11-03 11:04 ` Konstantin Khlebnikov
2019-10-23 20:18 ` [RFC PATCH 08/10] pipe: Rearrange sequence in pipe_write() to preallocate slot " David Howells
2019-10-23 20:18 ` [RFC PATCH 09/10] pipe: Remove redundant wakeup from pipe_write() " David Howells
2019-10-23 20:18 ` [RFC PATCH 10/10] pipe: Check for ring full inside of the spinlock in " David Howells
2019-10-24 10:32 ` [RFC PATCH 04/10] pipe: Use head and tail pointers for the ring, not cursor and length " David Howells
2019-10-24 13:14 ` [RFC PATCH 00/10] pipe: Notification queue preparation " Peter Zijlstra
2019-10-24 16:57 ` [RFC PATCH 11/10] pipe: Add fsync() support " David Howells
2019-10-24 21:29 ` Linus Torvalds
2019-10-25 8:34 ` David Howells
2019-10-27 15:22 ` Christoph Hellwig
2019-10-27 16:04 ` Konstantin Khlebnikov
2019-10-31 15:13 ` David Howells
2019-10-31 15:15 ` David Howells
2019-11-02 18:53 ` Linus Torvalds
2019-11-02 19:34 ` David Howells
2019-11-02 20:31 ` Andy Lutomirski
2019-11-02 22:03 ` Linus Torvalds
2019-11-02 22:09 ` Linus Torvalds
2019-11-02 22:30 ` Andy Lutomirski
2019-11-02 23:02 ` Linus Torvalds
2019-11-02 23:09 ` Linus Torvalds
2019-11-02 23:14 ` Andy Lutomirski
2019-11-03 12:02 ` Konstantin Khlebnikov [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=c8001e7c-8039-3efb-948b-482b88005660@yandex-team.ru \
--to=khlebnikov@yandex-team.ru \
--cc=christian@brauner.io \
--cc=dhowells@redhat.com \
--cc=gregkh@linuxfoundation.org \
--cc=keyrings@vger.kernel.org \
--cc=linux-api@vger.kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=linux@rasmusvillemoes.dk \
--cc=luto@kernel.org \
--cc=miklos@szeredi.hu \
--cc=nicolas.dichtel@6wind.com \
--cc=peterz@infradead.org \
--cc=raven@themaw.net \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).